Update README.md

README.md

@@ -8,23 +8,39 @@ base_model:
 pipeline_tag: zero-shot-image-classification
 tags:
 - medical
+datasets:
+- simwit/mimic-cxr
+- danjacobellis/chexpert
+- rajpurkarlab/ReXGradient-160K
+- BahaaEldin0/NIH-Chest-Xray-14
+- SampadKar/vindr-cxr
+metrics:
+- accuracy
+- bleu
 ---
 # CheXficient
 
-CheXficient is a vision-language foundation model for chest X-ray (CXR) interpretation, developed to enhance both data- and computation-efficiency.
-It enables joint image-text representation learning and supports prompt-based zero-shot classification.
+[Paper](https://arxiv.org/abs/2602.22843) | [GitHub](https://github.com/cwangrun/CheXficient)
+
+CheXficient is a vision-language foundation model for chest X-ray (CXR) interpretation, designed to improve both **data efficiency** and **computational efficiency** during pretraining.
+
+Instead of scaling indiscriminately to ever-larger datasets, CheXficient adopts a principled data curation strategy to selectively prioritize informative training samples. 
+This approach demonstrates that active, structured data selection can serve as a cost-effective alternative to brute-force dataset enlargement.
+
+The model follows a dual-encoder architecture and supports prompt-based zero-shot classification via joint image-text representation learning.
 
-This repository provides a Hugging Face-compatible implementation for seamless integration into research workflows.
 
 ------------------------------------------------------------------------
 
 ## Model Overview
 
--   Architecture: Vision-Language dual encoder
--   Input: Chest X-ray image + text prompts
--   Output: Image-text similarity logits and embeddings
--   Framework: PyTorch + Hugging Face Transformers
--   Intended Use: Research in medical AI and multimodal learning
+- **Architecture:** Vision-language dual encoder  
+- **Image Backbone:** DINOv2 (base)  
+- **Text Backbone:** BioClinicalBERT  
+- **Input:** Chest X-ray image + text prompts  
+- **Output:** Image-text similarity logits and embeddings  
+- **Framework:** PyTorch + Hugging Face Transformers  
+- **Intended Use:** Research in medical AI and multimodal learning 
 
 ------------------------------------------------------------------------
 
@@ -88,21 +104,15 @@ print(probs)
 ```
 
 
-## Intended Use
-
--   Zero-shot CXR findings classification
--   Prompt-based disease detection
-
-
 ------------------------------------------------------------------------
 
 ## Citation
 
 ``` bibtex
 @article{chexficient2024,
-  title={CheXficient: Efficient Vision-Language Learning for Chest X-ray Understanding},
+  title={A data- and compute-efficient chest X-ray foundation model beyond aggressive scaling},
   author={...},
   journal={...},
-  year={2024}
+  year={2026}
 }
 ```